Our laboratory has shown in many previous studies that myocardial function in rats is affected by aging, specifically by prolonging isometric twitch duration. Biochemical correlates have been shown to change in parallel as well. We have, in previous studies, demonstrated that mild exercise which is not sufficient to induce a training effect in young animals is nonetheless able to reverse/retard cardiovascular aging when applied relatively late in the lifespan of the rat. We have subsequently attempted to define the limits of exercise in terms of type, duration, and age at which exercise is begun. The present study, utilized a swimming model begun at 5 weeks of age and continued up to 17 months of age. The contractile and biochemical parameters of the myocardium did not differ between the two groups. The absence of a training effect on heart weight or the ratio of heart weight to body weight precludes a definitive interpretation of these results. One possibility is that up to 17 months (and unlike the case for senescence) in order for physical exercise to alter myocardial properties, a training effect on heart weight or body weight must occur.